Major Depressive Disorder (MDD) disrupts the lives of millions of people each year and presents a substantial societal and economic burden. Despite prior research, the mechanisms underlying treatment response and relapse in MDD remain unclear. Several treatments for MDD are available, but establishing the best form of treatment can be a protracted trial and error process where some patients remain unresponsive. Advances in imaging technologies and in computational image analysis techniques continue to provide new and unique opportunities for elucidating the neurobiological bases of MDD and the neural processes associated with treatment success. To accelerate knowledge in this area, we propose to apply a leading-edge multimodal imaging approach to identify biomarkers indexing complementary aspects of treatment-induced brain plasticity focusing on fronto-limbic and striatal circuitry. Magnetic resonance imaging (MRI) will include structural, functional, diffusion and perfusion imaging and MR proton magnetic resonance spectroscopy (1HMRS), which jointly reflect brain chemistry, morphology, tissue architecture, resting state activity and blood flow, which is coupled to metabolism. Longitudinal and cross-sectional analyses will identify baseline factors and treatment-related changes in imaging biomarkers predictive of treatment outcome that may translate into the clinical setting to guide more effective treatment strategies. Electroconvulsive therapy (ECT), used to treat refractory depression, is an established and highly effective procedure eliciting a rapid response in eligible individuals. Since response occurs over a relatively short time period compared to psycho- or pharmacotherapy, ECT will be used as the treatment model to establish neurobiological correlates of therapeutic response. Patients with MDD will be followed prospectively to characterize the trajectories of ECT-related biological responses, which are expected to overlap those of other forms of treatment. Imaging will be performed at 4 time points: prior to the 1st ECT treatment, after the 2nd ECT session, 1 week after completion of the ECT index series and at 6- months post treatment when relapse will be determined. Clinical assessments will be made at two additional interval time points. Demographically similar control subjects will be imaged twice to allow estimation of the variance associated with serial assessments and to determine normalization of biomarkers in association with treatment success in MDD. The potential impact of the proposed research to science and health is large. New scientific leads may inform novel treatment approaches, identify individuals at risk for developing depression, elucidate disease-related genetic or endophenotypic factors, identify subpopulations of MDD patients who are more likely to benefit from a particular treatment, and may predictively identify patients at high risk for relapse thereby allowing for the use of alternate or more aggressive individualized treatment strategies. Multimodal longitudinal imaging in the context of the rapid clinical response to ECT is an innovative approach ideally suited for charting the trajectory of mental illness to determine where, when and how to intervene.